An unmodified subset of Cosmopedia v0.1 data filtered by keywords: k8s and kubernetes
Dataset splits
The splits are same as Cosmopedia v0.1
Dataset features
The dataset has the following features:
text: the synthetic generated content from Cosmopedia v0.1.
format: the style of text, this can for example be a textbook, a blogpost, a story.. It can also be inferred from the prompt.
audience: the target audience defined in the prompt
